Re: [xml] Refactoring of the SAX interface for namespace support.

[Daniel Veillard <veillard redhat com>]

On Wed, Aug 13, 2003 at 10:19:46PM +0100, Graham Bennett wrote:
> I like the idea of exposing more of the parser's data this way.  The
> only thing that I don't like is that there doesn't seem to be any clear
> separation between the end of the information for the start of the
> element and the start of the content.
>
> So if I get the following sequence of events:
>
> startElementNS
> namespace
> namespace
> attributeNS
> attributeNS
> -----------------
> startElementNS
> endElementNS
> -----------------
> endElementNS
>
> Then I only know that I have all the information for the start of the
> element once I get the next element callback.  The callback that tells
> me this could equally well be a comment, text or processing instruction,
> or am I missing something?

  Yup that's right, but following the exchange with Dave Beckett
we need to provide the namespaces as part of the startElementNS
callback. I simply suggest to also add the number of attributes
to come:
         StartElementNs(
              void *ctx,
              const xmlChar *localname;
              const xmlChar *prefix,
              const xmlChar *URI
              int nb_namespaces,          // number of namespace definitions
              const xmlChar **namespaces, // namespaces (prefix/URI pairs)
              int nb_attributes
              )

  that's something Dave also asked for. The order of the attributes also 
doesn't matter,

> >       there is another option even more disturbing from an API viewpoint:
> >         change name to simple const xmlChar * zero terminated to
> >     const xmlChar * with a lenght in bytes, like for the character
> >     callbacks.
> >       goal would be to minimize the number of string copies needed, this could
> >       be very effective for attribute values which operates on a non-bounded
> >       vocabulary. Minimizing the number of string allocated for tags can
> >       be done very easilly by the parser since the values pertains to a fixed
> >       vocabulary this is part of the enhancements I have long planned to do
> >       in libxml2.
>
> This would be very desirable from my point of view, as in my app I am
> always going to copy the string again anyway.

   Well I would limit this to attribute values (well characters() also
 work that way), for small strings it's unclear it's a big gain.

Daniel

-- 
Daniel Veillard      | Red Hat Network https://rhn.redhat.com/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/